Comparative Analysis of Machine Learning Algorithms in Automatic Identification and Extraction of Water Boundaries

Monitoring open water bodies accurately is important for assessing the role of ecosystem services in the context of human survival and climate change. There are many methods available for water body extraction based on remote sensing images, such as the normalized difference water index (NDWI), modi...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Aimin Li, Meng Fan, Guangduo Qin, Youcheng Xu, Hailong Wang
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	water extraction modified normalized difference water index (MNDWI) remote sensing machine learning algorithm Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999
Acceso en línea:	https://doaj.org/article/9e52ce183e274cc6b26481b946695a88
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:9e52ce183e274cc6b26481b946695a88
record_format	dspace
spelling	oai:doaj.org-article:9e52ce183e274cc6b26481b946695a882021-11-11T15:08:03ZComparative Analysis of Machine Learning Algorithms in Automatic Identification and Extraction of Water Boundaries10.3390/app1121100622076-3417https://doaj.org/article/9e52ce183e274cc6b26481b946695a882021-10-01T00:00:00Zhttps://www.mdpi.com/2076-3417/11/21/10062https://doaj.org/toc/2076-3417Monitoring open water bodies accurately is important for assessing the role of ecosystem services in the context of human survival and climate change. There are many methods available for water body extraction based on remote sensing images, such as the normalized difference water index (NDWI), modified NDWI (MNDWI), and machine learning algorithms. Based on Landsat-8 remote sensing images, this study focuses on the effects of six machine learning algorithms and three threshold methods used to extract water bodies, evaluates the transfer performance of models applied to remote sensing images in different periods, and compares the differences among these models. The results are as follows. (1) Various algorithms require different numbers of samples to reach their optimal consequence. The logistic regression algorithm requires a minimum of 110 samples. As the number of samples increases, the order of the optimal model is support vector machine, neural network, random forest, decision tree, and XGBoost. (2) The accuracy evaluation performance of each machine learning on the test set cannot represent the local area performance. (3) When these models are directly applied to remote sensing images in different periods, the AUC indicators of each machine learning algorithm for three regions all show a significant decline, with a decrease range of 0.33–66.52%, and the differences among the different algorithm performances in the three areas are obvious. Generally, the decision tree algorithm has good transfer performance among the machine learning algorithms with area under curve (AUC) indexes of 0.790, 0.518, and 0.697 in the three areas, respectively, and the average value is 0.668. The Otsu threshold algorithm is the optimal among threshold methods, with AUC indexes of 0.970, 0.617, and 0.908 in the three regions respectively and an average AUC of 0.832.Aimin LiMeng FanGuangduo QinYoucheng XuHailong WangMDPI AGarticlewater extractionmodified normalized difference water index (MNDWI)remote sensingmachine learning algorithmTechnologyTEngineering (General). Civil engineering (General)TA1-2040Biology (General)QH301-705.5PhysicsQC1-999ChemistryQD1-999ENApplied Sciences, Vol 11, Iss 10062, p 10062 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	water extraction modified normalized difference water index (MNDWI) remote sensing machine learning algorithm Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999
spellingShingle	water extraction modified normalized difference water index (MNDWI) remote sensing machine learning algorithm Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999 Aimin Li Meng Fan Guangduo Qin Youcheng Xu Hailong Wang Comparative Analysis of Machine Learning Algorithms in Automatic Identification and Extraction of Water Boundaries
description	Monitoring open water bodies accurately is important for assessing the role of ecosystem services in the context of human survival and climate change. There are many methods available for water body extraction based on remote sensing images, such as the normalized difference water index (NDWI), modified NDWI (MNDWI), and machine learning algorithms. Based on Landsat-8 remote sensing images, this study focuses on the effects of six machine learning algorithms and three threshold methods used to extract water bodies, evaluates the transfer performance of models applied to remote sensing images in different periods, and compares the differences among these models. The results are as follows. (1) Various algorithms require different numbers of samples to reach their optimal consequence. The logistic regression algorithm requires a minimum of 110 samples. As the number of samples increases, the order of the optimal model is support vector machine, neural network, random forest, decision tree, and XGBoost. (2) The accuracy evaluation performance of each machine learning on the test set cannot represent the local area performance. (3) When these models are directly applied to remote sensing images in different periods, the AUC indicators of each machine learning algorithm for three regions all show a significant decline, with a decrease range of 0.33–66.52%, and the differences among the different algorithm performances in the three areas are obvious. Generally, the decision tree algorithm has good transfer performance among the machine learning algorithms with area under curve (AUC) indexes of 0.790, 0.518, and 0.697 in the three areas, respectively, and the average value is 0.668. The Otsu threshold algorithm is the optimal among threshold methods, with AUC indexes of 0.970, 0.617, and 0.908 in the three regions respectively and an average AUC of 0.832.
format	article
author	Aimin Li Meng Fan Guangduo Qin Youcheng Xu Hailong Wang
author_facet	Aimin Li Meng Fan Guangduo Qin Youcheng Xu Hailong Wang
author_sort	Aimin Li
title	Comparative Analysis of Machine Learning Algorithms in Automatic Identification and Extraction of Water Boundaries
title_short	Comparative Analysis of Machine Learning Algorithms in Automatic Identification and Extraction of Water Boundaries
title_full	Comparative Analysis of Machine Learning Algorithms in Automatic Identification and Extraction of Water Boundaries
title_fullStr	Comparative Analysis of Machine Learning Algorithms in Automatic Identification and Extraction of Water Boundaries
title_full_unstemmed	Comparative Analysis of Machine Learning Algorithms in Automatic Identification and Extraction of Water Boundaries
title_sort	comparative analysis of machine learning algorithms in automatic identification and extraction of water boundaries
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/9e52ce183e274cc6b26481b946695a88
work_keys_str_mv	AT aiminli comparativeanalysisofmachinelearningalgorithmsinautomaticidentificationandextractionofwaterboundaries AT mengfan comparativeanalysisofmachinelearningalgorithmsinautomaticidentificationandextractionofwaterboundaries AT guangduoqin comparativeanalysisofmachinelearningalgorithmsinautomaticidentificationandextractionofwaterboundaries AT youchengxu comparativeanalysisofmachinelearningalgorithmsinautomaticidentificationandextractionofwaterboundaries AT hailongwang comparativeanalysisofmachinelearningalgorithmsinautomaticidentificationandextractionofwaterboundaries
_version_	1718437161119776768

Comparative Analysis of Machine Learning Algorithms in Automatic Identification and Extraction of Water Boundaries

Ejemplares similares